检索结果-内蒙古大学图书馆

Research on intelligent access control technology of face recognition model based on parameter sharing and dense connection

引用

International Journal of Computational Intelligence Studies 2023年第3-4期12卷 255-270页

作者： Xu, Yonghua School of Computer Engineering Jinling Institute of Technology Jiangsu Nanjing211169 China Jiangsu Key Laboratory of Data Science and Smart Software Jinling Institute of Technology Nanjing211169 China

This study proposes a laboratory intelligent facial recognition system based on improved CNN, which significantly improves the accuracy of facial recognition by optimising the portrait recognition algorithm, improving CNN calculation and large parameter scale, and utilising perspective projection to improve portrait effect and sample utilisation. The experimental results show that the recognition rate has been improved by 10%, the CPU usage rate is less than 100%, and the model parameters have been reduced by more than 95%. This system can effectively and accurately recognise faces, making it suitable for embedded facial recognition devices. Copyright © 2023 Inderscience Enterprises Ltd.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

SPHASE: Multi-Modal and Multi-Branch Surgical Phase Segmentation Framework based on Temporal Convolutional Network

SPHASE: Multi-Modal and Multi-Branch Surgical Phase Segmenta...

引用

2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

作者： Long, Jun Hong, Junkun Wang, Zidong Chen, Tingxuan Chen, Yunfei Yang, Liu Central South University Big Data Institution Changsha China Central South University School of Computer Science and Engineering Changsha China

ISBN: (纸本)9798350337488

Surgical phase segmentation plays an important role in computer-assisted surgery systems, aiming to recognize what step or what action is operating in the video frame. Existing methods focus on improving the accuracy and precision of video segmentation, but ignore semantic consistency and temporal continuity of video frames in the intra-phase, which is necessary to apply in realistic computer-assisted equipment. Meanwhile, recent works almost extract long-term dependencies by Temporal Convolutional Network, but we heed high layers in TCN lose fine-grained information for detecting surgical steps and further affect phase segmentation task. To address these problems, we propose a Surgical Phase Segmentation Framework (SPHASE) which contains a multimodal feature fusion process and follows a multi-branch predictor. Moreover, we design a multimodal feature fusion mechanism when aggregate optical flow feature and I3D feature. The extensive experiments on AutoLaparo, Cholec80, and M2CAI2016 datasets demonstrate our method outperforms the state-of-the-art method by a large margin, especially in the JACC metric, which means SPHASE is more applicable in the surgical operating room. © 2023 IEEE.

关键词： attention mechanism multimodal feature fusion surgical phase segmentation temporal convolutional network

来源：评论

学校读者我要写书评

暂无评论

VRMDA: Verifiable and Robust Multi-subset data Aggregation Scheme in IoT

VRMDA: Verifiable and Robust Multi-subset Data Aggregation ...

引用

7th International Conference on Attacks and Defenses for Internet-of-Things, ADIoT 2024

作者： Li, Jianying Zhou, Shuo Liu, Yining School of Computer and Information Security Guilin University of Electronic Technology Guilin541004 China School of Data Science and Artificial Intelligence Wenzhou University of Technology Zhejiang Wenzhou325035 China

ISBN: (纸本)9783031855924

data aggregation masks individual entries by summing values, thus preventing the disclosure of private information. Multi-subset data aggregation divides subsets according to the range of data and collects the aggregated values of each subset. This provides the data control center with granular data. If data aggregation loses its verifiability, it becomes impossible to confirm the authenticity of messages or detect any tampering. This vulnerability could lead to an increased incidence of data tampering attacks across the system. Furthermore, a centralized control center is prone to disruptions due to natural disasters or malicious attacks, creating a single point of failure that may cause service interruptions and data loss. This could result in service interruptions and data loss. Therefore, we propose a robust and verifiable multi-subset data aggregation scheme. In this scheme, ciphertext information is transmitted to multiple control centers in the form of shards, and the aggregation results are recovered through cooperation between control centers. The security analysis demonstrates the protocol could resist collusion attacks and single points of failure. Experimental analysis indicates that the scheme is feasible. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： data aggregation

来源：评论

学校读者我要写书评

暂无评论

Question Embedding on Weighted Heterogeneous Information Network for Knowledge Tracing

引用

ACM Transactions on Knowledge Discovery from data 2024年第1期19卷 1-28页

作者： Sun, Jianwen Du, Shangheng Zhou, Jianpeng Yuan, Xin Shen, Xiaoxuan Liang, Ruxia National Engineering Research Center of Educational Big Data Central China Normal University Wuhan China Faculty of Artificial Intelligence in Education Central China Normal University Wuhan China Shanghai Institute of Artificial Intelligence for Education East China Normal University Shanghai China School of Computer Science and Technology East China Normal University Shanghai China School of Computer Science and Engineering Sun Yat-Sen University Guangzhou China

Knowledge Tracing (KT) aims to predict students' future performance on answering questions based on their historical exercise sequences. To alleviate the problem of data sparsity in KT, recent works have introduced auxiliary information to mine question similarity, resulting in the enhancement of question embeddings. Nonetheless, there remains a gap in developing an approach that effectively incorporates various forms of auxiliary information, including relational information (e.g., question-student, question-skill relation), relationship attributes (e.g., correctness indicating a student's performance on a question), and node attributes (e.g., student ability). To tackle this challenge, the Similarity-enhanced Question Embedding (SimQE) method for KT is proposed, with its central feature being the utilization of weighted and attributed meta-paths for extracting question similarity. To capture multi-dimensional question similarity semantics by integrating multiple relations, various meta-paths are constructed for learning question embeddings separately. These embeddings, each encoding different similarity semantics, are then fused to serve the task of KT. To capture finer-grained similarity by leveraging the relationship attributes and node attributes on the meta-paths, the biased random walk algorithm is designed. In addition, the auxiliary node generation method is proposed to capture high-order question similarity. Finally, extensive experiments conducted on six datasets demonstrate that SimQE performs the best among 10 representative question embedding methods. Furthermore, SimQE proves to be more effective in alleviating the problem of data sparsity. © 2024 Copyright held by the owner/author(s).

关键词： Question answering

来源：评论

学校读者我要写书评

暂无评论

Akane: Perplexity-Guided Time Series data Cleaning

引用

Proceedings of the ACM on Management of data 2024年第3期2卷 1-26页

作者： Xiaoyu Han Haoran Xiong Zhenying He Peng Wang Chen Wang X. Sean Wang School of Computer Science Fudan University Shanghai China Shanghai Key Laboratory of Data Science School of Computer Science Fudan University Shanghai China National Engineering Research Center for Big Data Software EIRI Tsinghua University Beijing China

Dirty data are prevalent in time series, such as energy consumption or stock data. Existing data cleaning algorithms present shortcomings in dirty data identification and unsatisfactory cleaning decisions. To handle these drawbacks, we leverage inherent recurrent patterns in time series, analogize them as fixed combinations in textual data, and incorporate the concept of perplexity. The cleaning problem is thus transformed to minimize the perplexity of the time series under a given cleaning cost, and we design a four-phase algorithmic framework to tackle this problem. To ensure the framework's feasibility, we also conduct a brief analysis of the impact of dirty data and devise an automatic budget selection strategy. Moreover, to make it more generic, we additionally introduce advanced solutions, including an ameliorative probability calculation method grounded in the homomorphic pattern aggregation and a greedy-based heuristic algorithm for resource savings. Experiments on 12 real-world datasets demonstrate the superiority of our methods.

关键词： data cleaning perplexity recurrent patterns time series

来源：评论

学校读者我要写书评

暂无评论

Knowledge-Enhanced Difference-Aware Clinical Time Series Representation Learning for Diagnosis Prediction

Knowledge-Enhanced Difference-Aware Clinical Time Series Rep...

引用

2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

作者： An, Ying Shi, Yinghong Guo, Lin Sheng, Yu Chen, Xianlai Central South University Big Data Institute Changsha China Central South University School of Computer Science and Engineering Changsha China

ISBN: (纸本)9798350337488

Predicting future health status based on historical patient visits is one of the essential tasks in healthcare. Many existing approaches attempt to enhance the representation learning capability of models by incorporating relevant medical knowledge, but their effectiveness is severely affected by the incompleteness and noise of the knowledge graphs. Moreover, due to the inability to capture temporal features at a fine-grained level, most existing methods also have limitations in learning the temporal development of patients' health status. To address these issues, we propose a Knowledge-Enhanced Difference-Aware clinical time series representation learning model (KEDA) for diagnosis prediction. In this model, we first combine the medical ontology graph and co-occurrence graph, and use hierarchical graph convolution and contrastive learning methods to enhance the semantic representation of medical entities. After that, a task-specific difference-aware temporal module is designed to improve the accuracy of patient representation, which adds two novel gated units in the original GRU to fuse multi-type clinical information based on the relationship between different types of data and prediction tasks and capture fine-grained temporal evolution of patient health status. We validate our model on two publicly available datasets, and the experimental results demonstrate that KEDA outperforms the state-of-the-art methods. © 2023 IEEE.

关键词： diagnosis prediction Electronic Medical Record fine-grained temporal feature knowledge graph multi-type fusion

来源：评论

学校读者我要写书评

暂无评论

A Lyapunov's Direct Method for Coorperative Output Regulation of Nonlinear Multi-Agent Systems 42

A Lyapunov's Direct Method for Coorperative Output Regulatio...

引用

42nd Chinese Control Conference, CCC 2023

作者： Wei, Songlin Su, Youfeng School of Mathematics and Statistics Fuzhou University Fuzhou350116 China College of Computer and Data Science Fuzhou University Fuzhou350116 China

ISBN: (纸本)9789887581543

The cooperative global robust output regulation problem for a class of nonlinear uncertain multi-agent systems with dynamic uncertainty has been approached by some distributed state feedback control law, however this method does not produce an explicit Lyapunov function for the closed-loop system. In this paper, we develop a Lyapunov's direct method to solving the global robust output regulation problem for the mentioned systems and produce the corresponding controller. Moreover, the Lyapunov function for the closed-loop system will be a superposition of those of individual subsystems. © 2023 Technical Committee on Control Theory, Chinese Association of Automation.

关键词： Lyapunov methods

来源：评论

学校读者我要写书评

暂无评论

Target Threat Assessment Method based on RCS Feature Classifier and TOPSIS 15

Target Threat Assessment Method based on RCS Feature Classif...

引用

15th International Conference on Microwave and Millimeter Wave Technology, ICMMT 2023

作者： Shan, Yujie Zhang, Jinhuan Dong, Jian Long, Jun Central South University School of Computer Science and Engineering Changsha China Central South University Big Data Institute Changsha China

ISBN: (纸本)9798350338874

Target threat assessment involves many uncertainties and is a tactical decision assessment problem with incomplete and uncertain information. In traditional target threat assessment, only the influence of the target's state attributes on its threat level is considered, while the target's type is ignored. To address this issue, radar target recognition is incorporated into the threat assessment process, and a method based on radar target recognition is proposed to improve the accuracy of threat assessment by incorporating Radar Cross Section (RCS) feature classifier into a Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). Using simulations, the practicability and viability of the above method are examined. © 2023 IEEE.

关键词： Radar target recognition

来源：评论

学校读者我要写书评

暂无评论

Natural Language Query and Control Interface for database Using Afghan Language 16

Natural Language Query and Control Interface for Database Us...

引用

16th International Conference on INnovations in Intelligent SysTems and Applications, INISTA 2022

作者： Karimi, Sadullah Rasel, Annajiat Alim Abdullah, Matin Saad Brac University School of Data and Sciences Department of Computer Science and Engineering Dhaka Bangladesh

ISBN: (数字)9781665498104

ISBN: (纸本)9781665498104

The Afghan language, or Persian language, is one of the most widely used languages, with up to 110 million speakers worldwide. It is used in countries like Afghanistan, Azerbaijan, Iran, Iraq, Russia, Tajikistan, Turkmenistan, Uzbekistan, etc. The Persian language has insufficient resources and needs higher quality lexicon translation. The proposed Afghan Language Query and Control Interface (ALQCI) is based on NLIDB Approach. The Afghan language is used in this system to work with user-chosen data from a database. Retrieving information from the database requires knowledge of SQL Query Language or a well-designed interface. It is convenient for domain-expert users to retrieve data from the database. However, it may be challenging for non-expert users to access the database through SQL Query in the absence of a flexible user interface. This system helps overcome the challenge for those who speak Persian worldwide. The system may supplement or substitute less friendly user interfaces (UI). The system generates Structured Query Language (SQL) when a user inputs a spoken natural language query in the Persian language. The system works by allowing a user to enter a query in Persian, which the system then translates into a SQL query using the word mapping technique. The SQL is generated through semantic analysis. Afterwards, the generated SQL is used to extract knowledge using the Afghan language from a database. © 2022 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Lightweight Weight Update for Convolutional Neural Networks 1

引用

12th International Congress on Big data, Bigdata 2023

作者： Wang, Feipeng Ben, Kerong Zhang, Xian Yang, Meini College of Electronic Engineering Naval University of Engineering Wuhan China School of Computer and Big Data Science Jiujiang University Jiujiang China

ISBN: (数字)9783031447259

ISBN: (纸本)9783031447242

Convolutional neural networks are usually composed of convolutional layers and pooling layers. Pooling operations effectively control the weight update of convolutional neural networks. The existing pooling operations result in a large number of weight parameter updates of convolutional neural networks, causing large memory usage. In this paper, a pooling operation called ApproxM is proposed to address this problem. The proposed pooling operation is a simple and similar to median pooling. It takes the mean of multiple values near the median as the pooling result, and CNNs only update the weights of these values during the back propagation. Finally, extensive experiments on benchmark datasets demonstrate that the proposed pooling operation achieves top-1 results of 92.65% and 68.24% and top-5 results of 99.84% and 91.31% model test accuracy on Cifar-10 and Cifar-100 based on ResNet-20, respectively, and the corresponding number of weight updates in an 8 × 8 pool is 4, which is better than other pooling techniques in the experiments. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Image recognition

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：